Advanced Question Thrust values of ship with other ships docked

[T1234]

When I add thrust values to my explorer addon and launch orbiter, the rotation and translation drastically reduces when I dock another ship to it and farther reduces when more ships are docked. Is this a bug or is there a way to ignore the mass of docked ships by inserting the correct piece of code.? For Orbiter 2016.

 

[Thorsten]

Not sure if I understand the question - but are you asking for how to disable the effect of inertia in simulated physics here? Because that's what gives you less acceleration for the same thrust with more mass.

 

[T1234]

The ship addon has huge mass and huge thrust,its an interplanetary vessel but the reduction in rcs and translation thrust is disproportionately large,so large it hardly moves even with a low mass like shuttlePB docked,but with no ship docked it is fine.

 

[ADSWNJ]

Calculate the effect from first principles and see if what you are seeing makes sense. Also think of the effect of the Center of Gravity post-docking, to see if the thrust is vectoring in a way different to you want, due to additional mass off the centerlines.

There's lots of ways to slew the vessel to do something non-realistic, but it's better to make it work without resorting to "suspend physics" trickery, if you can.

 

[T1234]

So if the rcs thrust points are set farther apart along the x axis,that may solve it?.The rcs is way more affected than the translation.

 

[ADSWNJ]

Maybe ... hard to tell without looking at the specifics. Keep going, friend!

 

[Thorsten]

Whether anything 'solves' it depends on what you see as a problem.

The rotational accelerations of a spacecraft are determined by the torque exerted by the thrusters and the inertia tensor. If you hard-dock ships together, the inertia tensor changes and the accelerations drop. That's not a problem as such, that's just sound physics.

If you need to design something that has high maneuverability even with multiple craft hard-docked for whatever reason and want to stay within physics, then you need to increase the torque exerted by the thrusters.

Since that is given by 'lever arm' times 'thrust', either increasing thrust or moving the thrusters farther from the center of gravity of the docked cluster will do if you want to stay within the realm of physics.

(Once you're willing to leave it, you can also manipulate the inertia tensor. Or not bother with solving equations of motions for attitude control at all and just directly change attitude - it's a computer simulation after all, it doesn't require you to go through any physics to get into some attitude).

 

[T1234]

Moving the thrust points away from the centre of gravity has solved it, but I do find it odd that a 1000 ton mass equivelant ship with ample thrust to rotate it
should be hindered by a 1 ton ship docked to it, other addons like deepstar don,t seem to behave this way,the behaviour is the same whether a ship is docked or not, maybie because it was made for orbiter 2010?.Anyway explorer is vastly improved and will update with new dll. Thanks for the replys.:thumbup:

 

[Thorsten]

but I do find it odd that a 1000 ton mass equivelant ship with ample thrust to rotate it
should be hindered by a 1 ton ship docked to it

Maybe it was unrealistic in the first place? If the inertia tensor of the original ship is little more than a pointmass at the origin, unrealistically weak thrust could already yank it around. But then docking (and providing a non-trivial inertia tensor) would uncover the problem rather efficiently - in which case you haven't actually fixed it, just hidden it again slightly better.

 

[francisdrake]

To find out why this seem unrealistic, you may try following:

First check that your ship mass is correct. E.g. that the mass is given in kg (not ton), and the values for dry mass and propellant are correct.

Then use the Shipedit tool to determine the Moment of Inertia. This is to rotation the same, what mass is for linear movement. The tool is in Orbitersdk/utils/Shipedit.exe.
Use the "File - Load Mesh" dialog to open your Explorer mesh.
Then use "Calc - Start / Continue MC integration". Let it run for some time.

Your ship is long and thin, so the interia tensor has two high values around x and y axis and a small one around z (the ship's longitundinal axis). Use the values of the main diagonal of the Inertia tensor. They are like 176813, 176551, 841.
(I tried it. These values are huge.)

Use these values to
SetPMI (_V(176813, 176551, 841)); // Prinzipal Moments of Inertia

Then test fly your ship. If it now is nearly unresponsive to rotational commands, your initial PMI setting was too low. You will need stronger RCS thrusters, and have them located farther away from the mass center to get proper response.

The positive side is, once you have corrected this, the compartively small mass of a ShuttlePB will no more influence your ship very much.

 

[T1234]

I found that moving the thrust points father apart has made the biggest difference.I am starting to understand tensors more.:thumbup: